Point process-based Monte Carlo estimation

DOI10.1007/S11222-015-9617-YzbMATH Open1505.62414arXiv1412.6368OpenAlexW2963963665MaRDI QIDQ517403FDOQ517403

Authors: Clément Walter 

Publication date: 23 March 2017

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: This paper addresses the issue of estimating the expectation of a real-valued random variable of the form $X=g(mathbfU)$ where $g$ is a deterministic function and $mathbfU$ can be a random finite- or infinite-dimensional vector. Using recent results on rare event simulation, we propose a unified framework for dealing with both probability and mean estimation for such random variables, emph{i.e.} linking algorithms such as Tootsie Pop Algorithm (TPA) or Last Particle Algorithm with nested sampling. Especially, it extends nested sampling as follows: first the random variable $X$ does not need to be bounded any more: it gives the principle of an ideal estimator with an infinite number of terms that is unbiased and always better than a classical Monte Carlo estimator -- in particular it has a finite variance as soon as there exists $kinmathbbR>1$ such that $operatornameE[X^k]<infty$. Moreover we address the issue of nested sampling termination and show that a random truncation of the sum can preserve unbiasedness while increasing the variance only by a factor up to 2 compared to the ideal case. We also build an unbiased estimator with fixed computational budget which supports a Central Limit Theorem and discuss parallel implementation of nested sampling, which can dramatically reduce its computational cost. Finally we extensively study the case where $X$ is heavy-tailed.

Full work available at URL: https://arxiv.org/abs/1412.6368

Recommendations

• Properties of nested sampling
• Computing the variance of a conditional expectation via non-nested Monte Carlo
• Diffusive nested sampling
• Unbiased approximations of products of expectations
• A general method for debiasing a Monte Carlo estimator

zbMATH Keywords

heavy tailscentral limit theoremnested samplingrare event simulationlast particle algorithm

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Monte Carlo methods (65C05) Statistics of extreme values; tail inference (62G32)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• On parallel implementation of sequential Monte Carlo methods: the island particle model
• On the Lambert \(w\) function
• An overview and open research topics in statistics of univariate extremes
• Properties of nested sampling
• Multilevel Monte Carlo Path Simulation
• Nested sampling for general Bayesian computation
• Adaptive Multilevel Splitting for Rare Event Analysis
• Simulation and estimation of extreme quantiles and extreme probabilities
• Sequential Monte Carlo for rare event estimation
• Exact sampling with coupled Markov chains and applications to statistical mechanics
• Efficient Monte Carlo simulation via the generalized splitting method
• Importance Sampling for Stochastic Simulations
• Estimating the mean of heavy-tailed distributions
• A combined splitting-cross entropy method for rare-event probability estimation of queueing networks
• Title not available (Why is that?)
• Title not available (Why is that?)
• Estimating the mean of a heavy tailed distribution
• Estimating the conditional tail expectation in the case of heavy-tailed losses
• Unbiased estimation with square root convergence for SDE models
• The Asymptotic Efficiency of Simulation Estimators
• A general method for debiasing a Monte Carlo estimator
• Random Construction of Interpolating Sets for High-Dimensional Integration
• Diffusive nested sampling

Cited In (7)

• Dynamic nested sampling: an improved algorithm for parameter estimation and evidence calculation
• Unbiased estimators and multilevel Monte Carlo
• Nested sampling methods
• A randomized multi-index sequential Monte Carlo method
• Computing the variance of a conditional expectation via non-nested Monte Carlo
• Nested sampling for uncertainty quantification and rare event estimation
• Unbiased approximations of products of expectations

This page was built for publication: Point process-based Monte Carlo estimation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q517403)